Unbalanced Fault

Unbalanced or unsymmetrical faults are very common in distribution systems, this type of fault leads to unequal current and unequal phase shift in a three phase system. There are three types of unbalanced faults namely single line to ground fault (LG), line to line faults (LL) and double line to ground fault (LLG). Impact of EV charging on each of these faults are discussed below.

(a) Single line to ground faults (LG)

Single line to ground faults are the most common fault in power system. As per the fault statistics of overhead lines [115], 80 % of the faults that occur in power system are single line to ground faults. In this work, a line to ground fault in phase A is applied at 0.33 s between node 888 and node 890 of the distribution feeder and the fault is cleared from the system at 0.42 s. Both case 1 i.e. with charging load connected at node 890 and case 2 with charging load replaced by equivalent constant power load are simulated.

Figure 4.2: Voltage of faulted phase A for LG fault

Fig. 4.2(a) gives the voltage of faulted phase A at node 890 for the period of fault and fault recovery for PEV load case and Fig. 4.2(b) shows the voltage of faulted phase A at node 890 during the period of fault and fault recovery for equiva- lent load case. As can be seen in Fig. 4.2(a) for the case 1 with PEV load connected at node 890, a transient voltage of high frequency having peak magnitude of approx- imately half of the steady-state peak voltage is observed at the time of reconnection of phase A to the system as compared to the equivalent constant power load case 2.

Once the phase A is reconnected back to the distribution system the voltage regains its steady-state value. This transient voltage appears due to continuous switching of charging system of PEV load connected to the distribution node.

Figure 4.3: Phase B current for LG fault

Fig. 4.3 gives phase B current at node 890 during fault condition and recovery period. Fault current is an important fault parameter which determines the severity of fault and affects the function of protective device in the system. As can be seen from Fig. 4.3(a) for case 1, with charging vehicle load connected to the node 890,

the phase B current at the time of fault shows the peak magnitude of 3 times the magnitude of steady state current. An increase in magnitude is also observed in both phase C current but major increase is observed in the phase B current due to mutual coupling of the distribution lines which is shown in Fig. 4.3(a). For case 2 with charging vehicles load replaced by equivalent constant power load the effect is not significant and the transient in the current during the fault is much less pronounced as compared to case 1.

Figure 4.4: Phase B reactive power for LG fault

Another significant change is observed in reactive power requirement in phase B during fault condition at node 890. For PEV load connected case the reactive power of phase B shows a sudden jump of approximately 21 times from 0.31 kVAR before fault to 6.6 kVAR at the time of fault and then settles to a value of 0.45 kVAR shown in Fig. 4.4(a). The sudden rise in reactive power observed in phase B is nearly 2.5 times the reactive power rise during fault with equivalent constant power load connected to node 890 as can be seen in Fig. 4.4(a) and Fig. 4.4(b). This

increase in reactive power requirement can have undesired effects like low voltages, increased system losses and equipment heating.

(b) Double line to ground faults (LLG)

Double line to ground faults are considered less severe as compared to the three phase fault and single line to ground faults on the overhead lines. In this work a double line to ground fault is simulated with phase A and phase B as the faulted phases in the line section connecting node 888 and node 890. The fault is applied at 0.33 s into the simulation and removed at 0.42 s. The simulation results are displayed in Fig. 4.5, in this fig (a), fig.(c) and fig.(e) shows results for PEV load case and fig (b), fig.(d) and fig.(f) display results for equivalent load case respectively. The voltage, current and the reactive power demand at node 890 are observed for each of the two cases. Fig. 4.5(a) and Fig. 4.5(b) show the voltage waveform at node 890 for the double line to ground fault for PEV load case and for equivalent load case respectively. As can be seen from the Fig. 4.5(a) with PEV load connected, a high frequency transient voltage appears during the breaker reclosing period whereas in equivalent load case shown in Fig. 4.5(b) no such transient voltage appears during breaker reclosing.

The effect on phase currents due to PEV charging load at node 890 is shown in Fig. 4.5(c). The fault current show sharp rise in magnitude from 15 A to 25 A in the first cycle of the fault and a switching transient current can be observed in the faulted phases during the fault and reclosing period. The fault current in equivalent load case is depicted in Fig. 4.5(d), here; the faulted phase current waveform does not show any sharp rise or transient during the fault time and the reclosing period.

With PEV charging load connected at node 890 the reactive power requirement show a sharp rise in healthy phase C from 386 VAR to 890 VAR as the fault strikes, which settles down to normal value after two cycles as shown in Fig. 4.5(e). The reactive power requirement for faulted phase A and phase B reduces to zero for the fault duration. Not much variation in reactive power requirement is observed in

Figure 4.5: Impact on voltage, current and reactive power for LLG fault at node 890

equivalent load case as shown in Fig. 4.5(f). For this case the reactive power for the healthy phase C varies between -1 VAR to 2 VAR for the fault duration.

(c) Line to line fault (LL)

Line to line faults are the second most common fault on overhead lines. This fault does not involve path to ground and are least severe in nature. A line to line fault case is simulated involving phase A and phase B in line section 888-890. The fault is applied at 0.33 s into the simulation and removed at 0.42 s. Fig. 4.6(a) and Fig. 4.6(b) shows three phase voltage for PEV load case and for equivalent load case respectively. The voltage, current and the reactive power demand at node 890 are observed. As can be seen from the Fig. 4.6(a) the effects are similar to double

line to ground fault case and a high frequency transient voltage appear during the breaker reclosing period with PEV load whereas in equivalent load case shown in Fig. 4.6(b) no such transient voltage appears during breaker reclosing.

The effect on phase current due to PEV charging load at node 890 is shown in Fig. 4.6(c). The fault current don’t show any significant rise in magnitude, only a switching transient current appears in the faulted phases for the fault duration and reclosing period. The fault current in equivalent load case is depicted in Fig. 4.6(d), as can be seen the faulted phase current waveform does not show any transient for fault duration and reclosing period.

0.3 0.35 0.4 0.45 0.5

time(s) (a)Three phase voltage -4

-2 0 2 4

Voltage(kV)

0.3 0.35 0.4 0.45 0.5

time(s) (c)Three phase current -15

-10 -5 0 5 10 15

Current(A)

0.3 0.35 0.4 0.45 0.5

time(s) (e)Reactive Power 0

0.2 0.4 0.6

kVAR _{Phase A}

Phase B Phase C

0.3 0.35 0.4 0.45 0.5

time(s) (b)Three phase voltage -4

-2 0 2 4

Voltage(kV)

0.3 0.35 0.4 0.45 0.5

time(s) (d)Three phase current -15

-10 -5 0 5 10 15

Current(A)

0.3 0.35 0.4 0.45 0.5

time(s) (f)Reactive Power -4

-2 0 2 4 6

VAR

Phase A Phase B Phase C

For PEV load For equivalent load

Figure 4.6: Impact on voltage, current and reactive power for LL fault at node 890

The reactive power requirement at the faulted node is shown in Fig. 4.6(e) and Fig. 4.6(f) for case 1 and case 2 respectively, no considerable variation in reactive

power requirement is observed for the PEV case and the equivalent load case due to the fault. For the PEV load case the reactive power requirement for faulted phase A and phase B reduces to zero for the fault duration and for the healthy phase C reactive power remains almost constant for the whole fault period and after the fault is removed. Not much variation in reactive power requirement is observed in equivalent load case as well, reactive power for the healthy phase C varies between -3 VAR to 6 VAR for the fault duration.